The manual does not give any warnings about using "load data local" when applying changes via the binary log.
I am having trouble getting my head around how this would work. IE if you update a table via "load data local" then how can you replicate these changes using the binary log, because when you try to run the SQL statements to replicate the changes of the table, you probably won't have that file local on your computer. (ie the original load data local could have been done from any computer on the network or even through the Internet) Does this mean replication servers can't be EXACTLY kept up to date if you use "load data local"? I can also see possible issues even if you don't use local - for example: A job uses "load data infile" that loads some data from the server into the table, then as a part of the clean up process deletes the input file. Now if you wanted to reproduce these changes from the binary log (eg after a file restore) the SQL statement that tries to run the "load data infile" will fall over because the file is no longer on the server. I would therefore think using "load data infile" local or otherwise has HUGE implications when you are looking at a backup strategy. Am I missing some thing here? Are there any work arounds, or should it be said that if you want EXACT replication of data DON'T use "Load data infile"? Thanks Clyde England --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
